离散时变系统的自适应迭代学习控制  

Adaptive iterative learning control of discrete time-varying systems

作　　者：孙明轩[1] 余林江[1] 

机构地区：[1]浙江工业大学信息工程学院,浙江杭州310023

出　　处：《浙江工业大学学报》2013年第1期84-90,共7页Journal of Zhejiang University of Technology

基　　金：国家自然科学基金资助项目(60874041;61174034)

摘　　要：针对一类有限区间上重复运行的离散时变SISO系统,分别采用带饱和函数和死区修正的投影算法进行参数估计,提出自适应迭代学习控制方案.关键技术引理在分析离散自适应控制系统时起到了关键作用,文中把这一引理推广至迭代域,用于建立离散自适应迭代学习控制系统的稳定性和收敛性.理论证明,即使每次迭代存在初始偏差,跟踪误差沿着迭代轴仍能收敛于零,且闭环系统的所有信号有界;当存在外部扰动时,跟踪误差收敛于一邻域内,其半径为干扰的界.在直线伺服系统上的应用结果验证了所提出的学习控制方法的有效性.By using the iterative learning projection algorithm with dead-zone and saturation modifications, adaptive learning control schemes are presented for a class of discrete time-varying SISO systems. The Key Technical Lemma plays an important role for establishing the stability of discrete adaptive control systems. The results given in this paper can be considered an application of the Key Technique Lemma in iterative domain, by which stability and convergence of the discrete adaptive iterative learning control schemes is established. Theoretical analysis establishes boundedness of all the closed-loop signals, and convergence of the tracking error to zero along the iteration axis. When considering the external disturbances, the tracking error can be ensured to converge to a bounded region, whose size is equal to the bound of the disturbances. A linear motor servo system is taken as an example, and the experiment results are presented for demonstrating the effectiveness Of the proposed method.

关 键 词：自适应迭代学习控制 投影算法 饱和函数 死区修正 离散时变系统 

分 类 号：TP273.22[自动化与计算机技术—检测技术与自动化装置]